Chi-square goodness-of-fit test: An example

The American Medical Association compiles information on physicians and publishes its findings in Physician Characteristics and Distribution in the U.S. Physicians can be broadly classified into four categories: general practice, medical, surgical, and other. [The "medical" category includes internal medicine and pediatrics; the "surgical" category includes general surgery, obstetrics and gynecology, orthopedic surgery, and ophthalmology; the "other" category includes psychiatry, anesthesiology, pathology, and radiology.] Table 1 provides a percentage distribution for U.S. physicians in 1986 based on those four categories.

Table 1  Specialty distribution of U.S. Physicians, 1986

A random sample of 500 U.S. physicians currently in practice yields the frequency distribution shown in Table 2.

Table 2  Sample results for specialties of 500 randomly selected U.S. physicians currently practicing medicine

At the 5% significance level, do the data provide sufficient evidence to conclude that the current specialty distribution of U.S. physicians is different from the 1986 specialty distribution?

SOLUTION

We will apply the macro FITTEST to perform a chi-square goodness-of-fit test. To begin click here.

  Back to home page for ICTCM talk